Employing defaulted leases, this study divides the loss given default (LGD) into two parts. So far, LGD has been regarded as a holistic measure of risk. However, considering the specifics of leases, we distinguish between asset-related and miscellaneous revenues of the workout process in order to calculate component LGDs. We introduce a multi-step approach to estimate the overall LGD of leases, based on its economic composition. The performance is assessed out-of-sample and out-of-time. We find that our approach generates stable and accurate estimations. Moreover, using the estimated component LGDs, we obtain valuable information regarding the debt collection procedure that lead to monetary advantages for the lessor.

This study employs a dataset from three German leasing companies with 14,322 defaulted leasing contracts to analyze different approaches to estimating the loss given default (LGD). Using the historical average LGD and simple OLS-regression as benchmarks, we compare hybrid finite mixture models (FMMs), model trees and regression trees and we calculate the mean absolute error, root mean squared error, and the Theil inequality coefficient. The relative estimation accuracy of the methods depends, among other things, on the number of observations and whether in-sample or out-of-sample estimations are considered. The latter is decisive for proper risk management and is required for regulatory purposes. FMMs aim to reproduce the distribution of realized LGDs and, therefore, perform best with respect to in-sample estimations, but they show poor performance with respect to out-of-sample estimations. Model trees, by contrast, are more robust and outperform all other methods if the sample size is sufficiently large.